1-s2.0-s0305750x02001109-main

Global Foreign Direct Investment Flows:

The Role of Governance Infrastructure

STEVEN GLOBERMANWestern Washington University, Bellingham, USA

and

DANIEL SHAPIRO *

Simon Fraser University, Vancouver, BC, Canada

Summary. — It is widely argued that a country�s economic performance over time is determinedto a great extent by its political, institutional and legal environment. We refer to these institutionsand policies as the governance infrastructure of a country. We utilize newly developed indices toexamine the effects of governance infrastructure on both foreign direct investment (FDI) inflowsand outflows for a broad sample of developed and developing countries over 1995–97. In addition,we examine the role of other forms of infrastructure including human capital and the environment.The results clearly indicate that governance infrastructure is an important determinant of both FDIinflows and outflows. Investments in governance infrastructure not only attract capital, but alsocreate the conditions under which domestic multinational corporations emerge and invest abroad.It would appear that investments in governance infrastructure are subject to diminishing returns, sothat the benefits, in terms of inflows, are most pronounced for smaller and developing economies.� 2002 Elsevier Science Ltd. All rights reserved.

Key words — foreign direct investment, capital flows, multinational corporations, infrastructure,

governance

1. INTRODUCTION

It is widely argued that a country�s econo-mic performance over time is determined to agreat extent by its political, institutional andlegal environment (OECD, 2001). We refer tothese institutions and policies as the governanceinfrastructure of a country. The governanceinfrastructure of a country helps to defineits investment environment, and thus createsfavorable conditions for economic growth.Recent empirical evidence tends to confirm

the hypothesis that crosscountry differencesin growth and productivity are related to dif-ferences in governance infrastructure (Hall& Jones, 1999; Kaufmann, Kraay, & Zoido-Lobaton, 1999b; Keefer & Knack, 1997; Knack& Keefer, 1995; Roll & Talbott, 2001). 1 Be-cause the investment environment of a countryaffects both domestic and foreign investors, andbecause foreign direct investment (FDI) hasbeen shown to promote host country efficiency,

it is a natural extension of the literature toconsider the impact of governance infrastruc-ture on crosscountry differences in FDI flows. 2

Our paper therefore focuses on the linkagebetween governance infrastructure and FDIflows. 3

The potential relevance of governance toexplaining FDI flows across countries has beenindirectly suggested by Lucas (1990), who ad-dresses the question of why capital flows fromrich to poor countries do not take place in theworld economy until capital to labor ratiosand, hence, wages and capital returns, are

World Development Vol. 30, No. 11, pp. 1899–1919, 2002� 2002 Elsevier Science Ltd. All rights reserved

Printed in Great Britain0305-750X/02/$ - see front matter

PII: S0305-750X(02)00110-9www.elsevier.com/locate/worlddev

*The authors thank Industry Canada, MicroeconomicPolicy Branch for financial support, and Mark Wilson

for research assistance. They also thank two unidentified

reviewers and the editor for many helpful comments and

suggestions. Dr. Someshwar Rao was inspirational in

suggesting the original focus of the study. Final revision

accepted: 18 June 2002.

1899

equalized. He considers a number of possibleexplanations and rejects several prominent ex-planations on conceptual grounds, includingthe possibility that technological change makescapital substantially more productive in devel-oped countries. An explanation that he con-siders quite plausible are the efforts of hostcountry governments to appropriate economicrents associated with inward FDI through in-struments such as heavy taxation. He offers thisas a possible explanation for relatively low ratesof capital formation in developed countries inthe face of substantial factor price differencesbetween developed and developing countries.While Lucas identifies explicit policies that aretargeted at foreign investors, other governancepolicies that discourage domestic capital invest-ment may also be relevant factors influencingintercountry differences in economic perfor-mance. In general, however, he highlights anargument that capital flows cannot be predictedby looking exclusively at labor and capitalscarcity.There is a relatively extensive empirical lit-

erature focusing on the characteristics of loca-tions that seem to either attract or repel foreigninvestors. 4 While it seems plausible that FDIwill be attracted to regions characterized bymore favorable governance infrastructures, allother things constant, most of the relevant lit-erature has focused on economic determinantsof FDI inflows. It is, of course, true that theinternational business literature has acknowl-edged the importance of country-specific po-litical risk (Kobrin, 1976). As a consequence,empirical analyses of FDI now routinely in-clude some kind of variable to control for in-tercountry differences in the broad politicalenvironment (Altomonte, 2000; Bevan & Estrin,2000; Mody & Srinivasan, 1998; Morisset,2000; Stevens, 2000; Tuman & Emmert, 1999),albeit with somewhat mixed results (Dawson,1998). 5

It is difficult to generalize about the statisticalimpact of political governance attributes, inpart because the attributes are measured indifferent ways in different studies. Moreover,although many previous studies adopt mea-sures that are closely related to the idea ofgovernance infrastructure, there has as yet beenno systematic attempt to relate directly gover-nance infrastructure measures to FDI flows fora wide cross-section of countries. Nor has therebeen much discussion regarding the specificinfrastructure elements that are especially ro-bust determinants of FDI.

Our paper adds to the relevant literature inseveral ways. Most prominently, we utilizenewly developed indices to examine the effectsof governance infrastructure on both FDI in-flows and outflows for a broad sample of(at most) 144 developed and developing coun-tries over 1995–97. Specifically, we use thegovernance indices developed by Kaufmann,Kraay, and Zoido-Lobaton (1999a) to measuregovernance infrastructure. These six indices,described below, cover a broad range of insti-tutional and policy outcomes and are availablefor a large sample of countries. In particular,they include factors not commonly found inthe FDI literature, notably measures of the ruleof law, the regulatory environment, and graft.Our results clearly indicate that governanceinfrastructure is an important determinant ofboth FDI inflows and outflows. The resultssuggest that investments in governance infra-structure not only attract capital, but also cre-ate the conditions under which domesticmultinational companies (MNC�s) emerge andinvest abroad. It would appear that invest-ments in governance infrastructure are subjectto diminishing returns, so that the benefits, interms of inflows, are most pronounced forsmaller and developing economies.Governance infrastructure is not the only

infrastructure that can contribute to economicwell-being and create a favorable climate forFDI. Investments in human capital, physicalinfrastructure and the environment may also beimportant. In the context of FDI, the absenceof educated and healthy workers can pose asignificant deterrent to foreign entry. As in-creasing amounts of FDI becomes skill- andefficiency-seeking, access to an educated andskilled workforce becomes essential. There isevidence that a more highly educated populacedoes in fact attract FDI (Mody & Srinivasan,1998), but the role of health has to ourknowledge not been explored. Similarly, envi-ronmental regulation may increase the costs ofdoing business and thus deter FDI. On theother hand, a clean environment may be asso-ciated with a higher quality of life, and thusattract FDI. To date, there are only a limitednumber of studies linking environmental poli-cies to FDI (List, 2001; Smarzynska & Wei,2001 and Wheeler, 2001), with no consistentevidence of a race to the bottom with respect toenvironmental policies. That is, there is noconsistent evidence of a negative relationshipbetween FDI inflows and higher environmentalstandards.

WORLD DEVELOPMENT1900

In this study, we account for aspects of hu-man capital development and the environ-mental regime using the Human DevelopmentIndex (HDI) developed by the United Nations,and the Environmental Sustainability Index(ESI) developed jointly at Columbia Univer-sity, Yale University and the World EconomicForum. The HDI is a composite index createdby combining GDP/capita, an education out-come index and a health status index. The ESImeasures environmental sustainability using avariety of different measures. Our study con-trasts the linkages between FDI and HDI andESI to the linkage between FDI and gover-nance infrastructure.The study proceeds as follows. In Section 2,

we define governance infrastructure and com-pare and contrast our definition to other relatedconcepts. In Section 3, we operationalize ourmeasure of governance infrastructure, as wellas of indices measuring human capital, physicalinfrastructure and environmental sustainabil-ity. Section 4 describes our statistical model.Section 5 discusses our estimation techniqueand results. A summary and conclusions isprovided in Section 6.

2. GOVERNANCE INFRASTRUCTURE

Broadly speaking, governance infrastruc-ture comprises public institutions and policiescreated by governments as a framework foreconomic and social relations. We are mostconcerned with those elements of the gover-nance infrastructure that can affect the in-vestment decisions of MNCs. A ‘‘positive’’governance infrastructure would therefore in-clude: an effective, impartial and transparentlegal system that protects property and indi-vidual rights; public institutions that are stable,credible and honest; and government policiesthat favor free and open markets. These con-ditions encourage FDI, and presumably privatedomestic investment as well, by protecting pri-vately held assets from arbitrary direct or in-direct appropriation. In a related manner, thesame conditions encourage sunk cost invest-ments by MNCs that facilitate efficient opera-tion in host countries.As we use the term, governance infrastruc-

ture is similar to the notion of social infra-structure used by Hall and Jones (1999) in thatthe definition includes both institutions andpolicies. We prefer the term governance be-cause it is readily distinguishable from related

notions of physical infrastructure, social capitaland human capital.Governance infrastructure, so conceived, can

be contrasted with physical infrastructure andhuman capital. Physical (public) infrastruc-ture is conventionally thought to include in-vestments in the construction and maintenanceof communications, transportation and utilitynetworks. Human capital reflects less tangibleinvestments in people, mainly in the form ofeducation and health. To the extent that edu-cation and health are provided by governmentor influenced by public policy, human capitalmay be thought of as human infrastruc-ture. Indeed, Vining and Weimer (2001) defineinfrastructure broadly to include both humancapital and physical infrastructure on thegrounds that they both facilitate investmentand growth, and are subject to market failure.Governance infrastructure can also be dis-

tinguished from social capital. Social capitalrefers to the networks and shared values thatencourage social cooperation, trust and, possi-bly, economic growth (Knack & Keefer, 1997;OECD, 2001). Unlike much physical capitaland governance infrastructure, social capitalresides in social relationships. Indeed, to theextent that transactions rely on sanctions andtrust (Humphrey & Schmitz, 1998), one maythink of sanctions (legal recourse, regulation)as elements of governance infrastructure, whiletrust emerges from moral and social norms.Nevertheless, social capital and physical andgovernance infrastructure may overlap becausesocial capital can involve public organiza-tions such as schools or government agencies(OECD, 2001, chapter 3). It might also beaugmented by investments in physical andgovernance infrastructure, as well as humancapital. In this regard, there is some evidenceto suggest that the existence of social capital(trust) is ‘‘linked to better performance ofgovernment institutions, including publiclyprovided education’’ (Knack & Keefer, 1997,p. 1253).In fact, a measure of social capital was ex-

cluded from this study for two main reasons.One is that there is no consensus in the litera-ture as to the appropriate way to specify socialcapital in studies focusing on differences inperformance among organizations or geo-graphical regions. 6 Second, and related to thefirst, the relationship networks underlying so-cial capital can be formed in many differentways. One would presumably need to aggregatethe various forms of relationship networks into

GLOBAL FOREIGN DIRECT INVESTMENT FLOWS 1901

broader indices comparable to governance in-dices. In this regard, we are unaware of theexistence of reliably estimated ‘‘meta-indices’’of social capital for even a few of our samplecountries. While the exclusion of social capitalmight contribute to biased estimates of thecoefficients for included infrastructure vari-ables, to the extent that social capital is sys-tematically correlated with the latter, theliterature says little about whether social capitaland governance infrastructure are stronglycorrelated in either a positive or negative di-rection.Governance infrastructure is related to mea-

sures of country-specific risk commonly used inthe international business literature (Bevan &Estrin, 2000; Keefer & Knack, 1997; Mody &Srinivasan, 1998). Private rating agencies typi-cally determine these measures by assigningweights to various economic, political and in-stitutional factors that define the investmentenvironment. These factors are not conceptu-ally much different from those used to definegovernance infrastructure. Indeed, we showbelow that there is a very high statistical cor-relation between our measures of governanceinfrastructure and one commonly used measureof country-specific risk. Nevertheless, the mea-sure of governance infrastructure we use isarguably more comprehensive.

3. MEASURING GOVERNANCE ANDOTHER INFRASTRUCTURE

Governance infrastructure is measured inour study by the six governance indicators es-timated by Kaufmann et al. (1999a,b). Theseindices (which we will refer to as KKZL indi-ces) describe various aspects of the governancestructures of a broad cross-section of coun-tries, including measures of political instability,rule of law, graft, regulatory burden, voice andpolitical freedom, and government effective-ness. 7 The indices have been estimated (usingan unobserved components model) employing31 different qualitative indicators from 13 dif-ferent sources, including BERI, DRI/McGrawHill, the Heritage Foundation, the WorldBank, the World Economic Forum and theEconomist Intelligence Unit. Thus, they are ina sense meta-indices, encompassing many ofthe various measures used in previous studies.Aggregate indicators drawn from a variety ofsources should provide more precise measuresof governance than individual indicators. A

further advantage is that these measures areavailable for an unusually large sample ofcountries (between 145 and 158 countries). Forthese reasons, we believe that the KKZL indi-ces are superior to other indices that have beenused in empirical studies.A disadvantage is that the indicators are es-

timated, and thus subject to measurementerror. But, the magnitude of the measurementerrors can be estimated which facilitates inter-pretation of how informative each indicatoris about the broader concept of governance(Kaufmann et al., 1999a). In addition, the in-dices are highly correlated with each other suchthat it is very difficult to use them all in a singleequation (Table 1). We have therefore createdan aggregate measure estimated as the firstprincipal component of the six measures. Werefer to this aggregated governance infrastruc-ture index as GII.In order to control for both physical infra-

structure and human capital, we employ theHDI published by the United Nations. Thisindex is now available for 168 countries, al-though not for every year. HDI is derived fromthree subindices: GDP/population, educationalliteracy and enrolment, and life expectancy atbirth. Each of the subindices is also available.We have calculated the average value of HDIfor 1995 and 1997. The health and educa-tion components are direct measures of humancapital. The GDP/population component is ameasure of wealth that we use as a proxymeasure for the amount of physical infra-structure. 8

Because neither the HDI nor the KKZL in-dices directly measure environmental quality orenvironmental regulation, we also employ theESI, created by The World Economic Forum,in conjunction with Columbia and Yale Uni-versities. The ESI index is derived from 22factors that contribute to environmental sus-tainability including air quality, public healthand environmental regulation. The ESI indextherefore reflects environmental infrastructurein the form of policy choices made by govern-ments, as well as human capital reflected inpublic health conditions.We treat the HDI and ESI indices as mea-

sures of human capital and physical and envi-ronmental infrastructure, but they may alsomeasure development outcomes. As a conse-quence, the three indices (GII, HDI and ESI)may be related. In particular, effective gover-nance may be a determinant of developmentoutcomes, as measured by HDI or ESI. 9


Table 1. Correlation matrix: governance infrastructure and other measures n ¼ 144

Mean (sd) HDI GDPC EDUC LIFE GII VOICE INSTAB GOV REG LAW GRAFT ESI

HDIa 0.68 (0.19) 1.00GDPC 0.63 (0.25) 0.93 1.00EDUC 0.75 (0.18) 0.90 0.70 1.00LIFE 0.68 (0.18) 0.94 0.81 0.80 1.00GIIb 0.01 (0.96) 0.69 0.69 0.53 0.60 1.00VOICE 0.06 (0.93) 0.59 0.59 0.50 0.52 0.85 1.00NSTAB )0.02 (0.93) 0.64 0.66 0.52 0.58 0.88 0.67 1.00GOV )0.02 (0.88) 0.63 0.69 0.44 0.55 0.95 0.75 0.78 1.00REG 0.07 (0.78) 0.51 0.56 0.37 0.44 0.84 0.73 0.66 0.75 1.00LAW 0.04 (0.92) 0.69 0.75 0.51 0.60 0.94 0.69 0.87 0.88 0.72 1.00GRAFT )0.01 (0.90) 0.65 0.71 0.49 0.55 0.93 0.75 0.74 0.93 0.67 0.87 1.00ESIc 49.49 (11.3) 0.65 0.62 0.61 0.53 0.78 0.73 0.63 0.72 0.64 0.67 0.75 1.00

aHDI is the Human Development Index published by the United Nations Development Program, averaged for 1995 and 1997. HDI combines three measures, gdpper capita (GDPC), education, measured by a combination of adult literacy and the combined gross primary, secondary and tertiary enrolment (EDUC) and lifeexpectancy at birth (LIFE). Index range is 0.0–1.0.bGII is the first principal component of a series of governance indicators estimated by Kaufmann et al. (1999a) for the World Bank. The KKZL indices are themselvesestimated by aggregating a number of measures for 1997. VOICE (Voice and Accountability) includes measures of political and civil liberties as well as freedom of thepress. INSTAB (Political Instability and Violence) includes measures of political violence, terrorism and ethnic conflict. GOV (Government Effectiveness) includesmeasures of government efficiency. REG (Regulatory Burden) includes measures of the degree of regulation and market openness, including tariffs, and import, exportand FDI restrictions. LAW (Rule of Law) is a measure that includes costs of crime, contract enforcement, and property rights. GRAFT (Graft), includes measures ofcorruption. Indices range from )2.5 to 2.5.c ESI is the Environmental Sustainability Index, published by The Center for International Earth Science Information Network (CIESIN) at Columbia University, andcreated with the Yale University Center for Environmental Law and the World Economic Forum. The ESI is based on 22 factors that contribute to environmentalsustainability, such as air quality, public health and environmental regulation. Based on 2000 data. Index ranges from 0 to 100. The ESI index was available for only 114countries.

GLOBALFOREIGNDIRECTINVESTMENTFLOWS

1903

Nevertheless, we include these measures be-cause development outcomes are also rele-vant to any discussion of FDI flows. The FDIliterature suggests that host country wealth(normally measured by GDP/capita) is an im-portant determinant of FDI flows (Dunning,1993). Moreover, some recent evidence suggeststhat the location decisions of foreign investorsmay be influenced by quality of life variables, ofwhich GDP/capita is but one (Peterson, Mal-hota, & Wagner, 1999). Given that GDP/capitais not necessarily a good measure of well-beingor quality of life (OECD, 2001), the HDI andESI indices may serve as such measures andtherefore attract FDI.The means and correlation coefficients for

the main indices (GII, HDI and ESI), and theircomponents are presented in Table 1. 10 Allmeasures are quite highly correlated, but thewithin group values are typically higher thanthose between groups. In particular, the HDIand GII indices are highly correlated with theirindividual component measures. 11 It is, there-fore, inappropriate to include individual com-ponent measures in the estimating equation, asdoing so would provide little more informationthan is gained by including only the HDI andGII indices.The HDI and GII indices are correlated

ðr ¼ 0:69Þ, which is not surprising given thatthe HDI index likely measures both inputs andoutput. The ESI variable is the least correlatedwith any other measure, and it is the onlyvariable that explicitly accounts for environ-mental quality. It is not however, available foras large a sample of countries (122 in total, butonly 114 in our sample).We experimented with various combinations

of the KKZL and HDI subindices. For exam-ple, we created a human capital index that wasthe sum of the education and health subindicesof the HDI. This variable was still highly col-linear with HDI and the GDP/capita compo-nent of the index. Similarly, we created a newvariable from the KKZL indices that was thesum of the government efficiency, regulatoryburden, and legal system efficiency indices. Thisvariable was also highly collinear with GII, andwith the remaining KKZL indices.As noted earlier, the KKZL indices are esti-

mated and, therefore, possibly subject to mea-surement error. We attempted to assess theirreliability by comparing them to a measure ofpolitical risk published by Institutional InvestorMagazine. This measure is a composite indexderived from a variety of sub-measures, but its

components are not published. As noted above,the measure is often used in the FDI literature(Bevan & Estrin, 2000; Mody & Srinivasan,1998). The KKZL indices and their first prin-cipal component (GII) are all highly correlatedwith the Institutional Investor risk variable.For example, the correlation between the latterand GII is r ¼ 0:87. Thus, despite the possi-bility of measurement error, it would appearthat the KKZL measures are robust, at least inrelation to ‘‘expert’’ judgments of national po-litical environments.

4. MODELING FDI INFLOWS ANDOUTFLOWS

The basic question we seek to address iswhether governance infrastructure, as measuredby GII, affects FDI inflows and outflows acrosscountries. In doing so, we also consider theimpact of physical and environmental infra-structure, as well as human capital. In order toestimate the impacts of the variables of interest,we need to hold constant other potentially im-portant influences on FDI within the confines ofa parsimonious model. The model chosen toestimate FDI inflows is specified as Eq. (1).

Ln FDIit ¼ b0 þ b1Ln GDPit�1

þ b2Governance Infrastructure Index ðGIIÞitþ b3Human Development Index ðHDIÞit�1þ b4 Environmental Sustainability Index ðESIÞitþ interactive termsþ eit

ð1ÞGloberman and Shapiro (1999) have argued

that FDI inflows and outflows are symmetrical.The presumption is that capital outflows maybe stimulated by the same factors that encour-age capital inflows. Specifically, superior gov-ernance encourages inward FDI, as well asincreased capital investment more generally.Some of the successful firms created through thedomestic investment process may, in turn,invest abroad as world-class multinational com-panies. In effect, superior governance encour-ages capital investment and the expansion ofbusinesses that, in turn, are associated withincreases in inward and outward FDI. Accord-ingly, the same specification is also used to es-timate equations whose dependent variables areeither capital outflows (Ln FDO), or net capitalflows, defined as Ln (FDIit�FDOit). In the nextsubsection, we discuss in more detail how thestatistical model was chosen and specified.


(a) Control variables

A large number of variables have been con-sidered in the literature as possible determi-nants of inward FDI. 12 In fact, however,surprisingly few are consistently significantacross the broad set of empirical studies thathave been performed. One variable that isconsistently statistically significant is a measureof the host country�s size, usually identified bya measure of real gross domestic product(GDP). 13 The theoretical linkage between realGDP and location advantage is straightfor-ward. A larger market implies that distributioncosts will be lower when production and dis-tribution facilities are sited in that marketwhere, presumably, the bulk of a seller�s cus-tomers will be located. As a related point, aclustering of other producers in the large mar-ket may create or accentuate agglomerationeconomies that, in turn, lower costs for allproducers in that market. Contributing to therelevant agglomeration economies may be theavailability of highly specialized inputs thatcannot be found in smaller markets. 14

Other variables provide less consistent re-sults. As noted, GDP per capita is often em-ployed as a measure of how well-off consumersare in a country. The problem with the GDPper capita variable is that it is also an implicitmeasure of wage rates, since productivity levelsare highly correlated with wage rates, as well aswith GDP per capita. All other things constant,higher wage rates will discourage inward FDI.Similarly, relative wage rates will implicitly re-flect productivity differences among countries.Hence, they will not necessarily reflect differ-ences in unit labor costs that, in principle, arewhat they are meant to measure. Consequently,it is not surprising that GDP per capita andrelative wage rates are frequently either statis-tically insignificant or appear with the ‘‘wrong’’signs in FDI regression equations. 15

We followed the literature in selectingcontrol variables reflecting the openness ofthe economy ðimportsþ exports=GDPÞ, laborcosts (wages and salaries per employee in man-ufacturing), taxation (government tax revenue/GDP), exchange rate instability (measured bydummy variables classifying the country�s ex-change rate regime as fixed against the USdollar, fixed against some other currency,managed floating or free floating), 16 and threemeasures of physical infrastructure (Internethosts per 10,000 people; telephone mainlinesper 1,000; millions of kW-h of electricity gen-

erated/GDP). None of these control variableswas ever statistically significant in any specifi-cation estimated. Moreover, each was avail-able for a smaller sample than the variablesultimately included. As a consequence we donot include them in the final model as sum-marized in equation one. 17

It is unsurprising that some of these vari-ables were not found to affect FDI flows, de-spite some theoretical and empirical supportfor their relevance in the literature. The po-tential ambiguity of relative wage measureswas discussed earlier. With respect to tax dif-ferences, the conceptually appropriate measureto compare across countries is the marginaleffective tax rate. This rate differs from in-dustrial sector to sector, and it is extremelydifficult to measure (Chen, 2000). Broadermeasures (such as tax revenues/GDP) do notmeasure the impact of taxation at the margin.In addition, there is considerable intracountryvariation in tax rates within large countries,and simple averages may disguise the ability ofa particular region to attract FDI. Finally, anyaversion to high taxes might be mitigated bytheir link to the provision of infrastructurethat, in turn, is highly valued by internationalinvestors.The fact that we could find no link between

FDI flows and measures of physical infra-structure is at odds with the recent literature,which tends to find a positive and statisticallysignificant effect. 18 In our case, the problemwas multicollinearity between measures ofphysical infrastructure and measures of GDPor HDI (mainly the GDP/capita component).Larger and richer countries are characterizedby more physical infrastructure. For example,the correlation coefficient between telephonesper capita and HDI is r ¼ 0:94. When ourphysical infrastructure measures were regressedagainst FDI in the absence of GDP and HDI,they were statistically significant and positive.For this reason, HDI must be considered, as apractical matter, to measure both physical in-frastructure and human capital.Similarly, the openness of an economy,

measured by trade flows as a ratio of GDP, islikely related to a host country�s legal and po-litical framework that, in turn, is supportive ofbusiness investment. Although trade variableswere never significant, the regulatory burden(REG) index of GII is to a great degree ameasure of openness, since it includes mea-sures such as tariffs and other trade restric-tions, resulting in collinearity between the trade


measure and the GII index. In fact, our resultsindicate that open economies attract FDI.The relationship between FDI and the ex-

change rate is more complex. The relevant issueis whether greater volatility of exchange ratesdiscourages FDI. On the surface, it would seemto be the case, since risk-averse investors pre-sumably view such volatility as a direct cost (ifhedging is used to reduce the volatility) or anindirect cost (if risk is unhedged). But, to theextent that MNCs operate across a number ofexchange rates, the volatility of any one cur-rency might actually reduce the overall vola-tility of the MNC�s cash flow. This will be thecase, for example, if movements in that cur-rency are uncorrelated, or negatively corre-lated, with movements of other currencies inwhich the MNC operates. In this case, currencyvolatility might be largely offsetting for MNCsoperating across a ‘‘basket’’ of currencies. Inshort, it is theoretically unclear how tradeopenness and exchange rate volatility affectFDI flows, and our results may reflect thistheoretical ambiguity.

(b) The dependent variable

We measure FDI in terms of flows. To theextent that inward and outward FDI have beengoing on for a long time, recent and relativelylarge changes in FDI behavior may not be ap-parent if FDI stock figures are used. That is,changes in stocks on a year-to-year basis will bequite small when they occur against an abso-lutely large accumulated base value. As a result,it may be difficult to identify the empiricalfactors affecting FDI stock values given rel-atively small variations in the FDI stockdependent variable. Moreover, inward andoutward FDI behavior is more comprehen-sively measured for flows than for stocks.The data on both inflows and outflows were

obtained from the United Nations publica-tion, The World Investment Report (UNCTAD,1998, Annex B). The UN, in turn, obtainedmost of the data directly from the InternationalMonetary Fund�s computer tapes. In thosecases where a country did not report to theIMF, data were obtained from the UNCTADFDI/TNC database.

(c) Specifying the model

The model is specified such that both FDIflows and GDP are measured in logarithms,with the GDP coefficient measuring the elas-

ticity of FDI flows. Given its GDP level, acountry will be more or less attractive to for-eign investors depending upon the extent andnature of its infrastructure and quality of life.Alternative specifications to (1) were consid-

ered and tested. In particular, we estimatedmodels in which the dependent variable wasspecified as being the ratio of FDI (inflows oroutflows) to GDP, and the Ln GDP term wasdropped as an explanatory variable. This spec-ification was rejected because the depen-dent variable was typically clustered within anarrow range, and the limited variation pro-duced very unreliable parameter estimates andlow degrees of explanatory power when eitherordinary least squares (OLS) or Tobit estima-tion methods were employed. As an alternative,the logisitic transformation of the FDI/GDPratio was calculated and employed as the de-pendent variable. This specification producedresults that are similar to those reported below.Indeed, there is virtually no difference in termsof levels of significance of the explanatoryvariables, and none of our conclusions wouldchange as a consequence of using this alter-native specification. 19

In addition, we estimated models in whichthe dependent variable was specified as theproportion of total global FDI received by anycountry (PFDI), or the logistic transformationof PFDI. These measures were highly corre-lated with Ln FDI, suggesting some indiffer-ence as to the choice among them. Thus, theresults are in fact similar, regardless of how thedependent variable is measured, and so we onlypresent results based on the (natural) logarith-mic specification (with GDP also in naturallogs). This specification allows greater flexibil-ity in that it allows the elasticity of FDI withrespect to GDP to be estimated, and it permitsus to introduce lagged GDP as an explanatoryvariable.In terms of independent variables, we ar-

rived at the final specification by eliminating allcontrol variables that were not statisticallysignificant in preliminary estimations. In the(unreported) results, the GDP variable wasalways statistically significant; moreover, it ex-plains virtually all of the variance in the de-pendent variable. As noted above, none of theother control variables for which data wereavailable was ever statistically significant withthe exception of GDP/capita. Since GDP/capitais part of the HDI index, we control only forGDP (measured in logarithms) in our reportedresults. Standard F-tests indicate that this


model is preferred over ones that also includethe control variables discussed above.The simple specification described by equa-

tion (1), without interaction terms, was sub-jected to RESET specification tests (discussedbelow). When the specification failed the RE-SET test, we considered specifications in whichthe GII, HDI and ESI indices were interactedwith the Ln GDP term. When the inclusion ofthe interactive term or terms allowed the spec-ification to pass the RESET test, they are re-ported. To the extent possible, all independentvariables were lagged relative to the dependentvariable. The measurement of the variables isdiscussed in the next section.

(d) Data and measurement

The sources and measurement of all variablesis summarized in Table 2. We were able tomeasure most variables for a cross-section of144 countries. The ESI variable was available

for only 114 countries, while only 98 countriesrecorded FDI outflows. At the time the datawere collected, 1997 marked the last date forwhich FDI data were available. But, use of asingle year�s data on FDI flows can be mis-leading, particularly for small countries, wherea single transaction in a given year can createtemporary and possible large changes in re-corded FDI flows, including negative values. Inorder to minimize this possibility, we chose toaverage the FDI data over 1995–97. At thesame time, the GII measures were available foronly one year (1997), and the HDI indices werenot available for every year, thus limiting ourability to create a useful time-series panel.In fact, there is remarkable temporal stability

in most of the relevant variables employed inthis study. For example, FDI inflows in 1995and 1996 have a simple correlation coefficientof 0.75. The correlation coefficient for FDIinflows for 1996 and 1997 is 0.986. Out-ward FDI flows are also highly correlated on a

Table 2. Variables, definitions and data sources

Variable Definition Source

FDI FDI inflows in $US, averaged 1995–97 UN World Investment (UNCTAD, 1998)FDO FDI outflows in $US, averaged 1995–97 UN World Investment (UNCTAD, 1998)FDIN Net FDI flows (FDI inflows minus FDI outflows),

averaged 1995–97UN World Investment (UNCTAD, 1998)

GDP Real GDP in 1990 $US, average 1994–96 United Nations Statistical Yearbook(various years)

HDI Human Development Index, averaged 1995 and 1997.Index combines GDPCI, EDUCI AND LIFEI

UNDP (various years)

GDPCI GDP/capita index, measuring standard of living,averaged 1995 and 1997.


EDUCI Education index, combining adult literacy and primary,secondary and tertiary enrolment ratios, averaged1995 and 1997.


LIFEI Life expectancy at birth index, averaged 1995 and 1997. United Nations Development Programme,various years

GII First Principal Component of Governance Indices(LAW, INSTAB, REG, GOV, GRAFT, VOICE,developed by Kaufmann et al. (1999a)

Kaufmann et al. (1999a), data available at:http://www.worldbank.org/wbi/governance/datasets.htm#dataset

LAW Rule of Law Index, measures contract enforcement,property rights, theft and crime, etc.

Kaufmann et al. (1999a)

INSTAB Political Instability and Violence Index, measures armedconflict, social unrest, ethnic tensions, terrorist threats etc.


REG Regulatory Burden Index, measures governmentintervention, trade policy, capital restrictions etc.


GOV Government Effectiveness Index, measures red tape andbureaucracy, waste in government, public infrastructure etc.


GRAFT Graft and Corruption Index, measures corruptionamong public and private officials, extent of bribery etc.


VOICE Voice and Accountability Index, measures civil liberties,political rights, free press, fairness of legal system etc.


ESI Environmental Sustainability Index, 2000, measures thehealth of the environmental system

Available at:www.ciesin.columbia.edu/indicators/ESI


year-to-year basis. Specifically, the correlationcoefficient between outward FDI in 1995 and1996 is 0.965. For the years 1996 and 1997, thesimple correlation coefficient is 0.981.Key independent variables are also highly

correlated over the mid-1990s sample period.For example, the index of human development(HDI) created for 1995 has a simple correlationof 0.979 with the same index calculated for1997. The subindex measuring educationalattainment for 1995 has a simple correlationequal to 0.992 with the same index for 1997.Real GDP in 1995 has a correlation of 0.999with real GDP in 1997.In summary, values of relevant dependent

and independent variables in models of FDIbehavior change relatively slowly over time. 20

As a consequence, although adjustments tochanges in the relevant independent variablesdo not occur immediately, departures fromequilibria are arguably ordinarily small, at leastrelative to past FDI behavior. Discrete andsubstantial short-run departures from equilib-rium values for the dependent and independentvariables would presumably be associated withmuch lower year-to-year correlations of eachvariable than we identify for our sample, asdiscussed above. As a result, cross-section dis-tributions of the relevant variables may, for oursample time period, reasonably approximate asteady-state equilibrium. Given the highly cor-related values of the dependent variables acrossour sample of countries, the precise choice ofyear(s) for those variables does not seem acrucial issue.To the extent possible, we attempted to

measure the independent variables for a priorperiod. GDP is measured in US dollars andaveraged for 1994–96. The GDP variable waslagged both to allow for adjustment lags, andto reduce the potential for a bias created by thedependent variable (FDI) causing higher valuesof GDP. HDI and its subindices are averagedover the two years 1995 and 1997. GII and itssubindices were only available for 1997, but it isdoubtful that its value would change much overa relatively short period of time. The earliestyear for which ESI was available was 2000, butit is also doubtful whether this index changesmuch over time. Of course, it would be pref-erable to lag all explanatory variables (or atleast to test for the appropriate lags), but dataconstraints did not allow us to do so.We conclude that a cross-section sample of

countries should allow the identification of along-run relationship between FDI and infra-

structure attributes. Since the FDI values arealso highly correlated on a year-to-year basisfor the mid-1990s, as are the values of the in-dependent variables, it seems unnecessary toestimate different cross-sections over time or topool cross-sections over time. A single cross-section in which the relevant variables areaveraged over the sample period seems a suffi-ciently robust approach to modeling in thiscase.

(e) Multicollinearity

Despite the parsimony of the basic model,there is still a potential problem created byintercorrelations among the independent vari-ables of interest, i.e. real GDP and the mea-sures of infrastructure. In addition to thecorrelations among the infrastructure variablespresented in Table 1, Ln GDP is correlatedwith the HDI index ðr ¼ 0:92Þ, and also withGII ðr ¼ 0:69Þ. These correlations underscorethe potential difficulty in statistically identifyingthe influence of specific infrastructure measuresin FDI models that include conventional eco-nomic variables such as GDP.

5. ESTIMATION AND RESULTS

(a) The FDI model

The basic results for the FDI model arefound in Tables 3 and 4. Each reported equa-tion was estimated by OLS, with heterosked-astic-consistent standard errors. The dependentvariable is measured in logarithms, as is GDP.The other variables (HDI, GII, ESI) are indicesand are not transformed. We tested for speci-fication error through a series of RESET tests(one and two power). For the full sample (de-fined below), it was found that any equationwith GII alone always failed the RESET test(one power), but passed comfortably when GIIwas interacted with Ln GDP. Interactive termsinvolving HDI and ESI were never statisticallysignificant. The implication of this finding willbe discussed below. All other equations passedthe RESET tests.We were also concerned about measurement

error, particularly with respect to the gover-nance variables. These variables were estimatedby Kaufmann et al. (1999a), and therefore eachobservation has a standard error. For each ofthe KKZL variables, we took the ratio of thestandard error to the mean estimate, and in-


Table 3. Regression results, FDI inflows dependent variable is Ln FDI

All countries dependent variable is Ln FDI Developing and transition economies dependent variable is Ln FDI

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

Ln GDP 0.951�� 0.826�� 0.764�� 0.791�� 0.737�� 0.893�� 0.903�� 0.845�� 0.862�� 0.861��

(0.059)a (0.080) (0.062) (0.085) (0.070) (0.088) (0.093) (0.078) (0.107) (0.085)Human DevelopmentIndex (HDI)

)0.374(0.881)

0.874(0.872)

)0.328(0.850)

0.470(0.798)

Education Index (EDUC) 1.190�� 2.068�� 1.183�� 1. 703��

(0.556) (0.669) (0.570) (0.689)Governance InfrastructureIndex (GII)

2.083�� 1.525�� 0.969�� 0.569��

(0.735) (0.774) (0.219) (0.209)Regulation Index (REG) 1.101�� 1.076�� 1.080�� 1.043��

(0.156) (0.214) (0.173) (0.239)Environment Sustain-ability Index (ESI)

0.005 )0.021 0.023 )0.009(0.015) (0.014) (0.019) (0.020)

Ln�GDP��GII )0.124�� )0.090�(0.062) (0.049)

Constant )3.857�� )2.260�� )2.980�� )2.635�� )3.636�� )3.413�� )2.98�� )3.698�� )4.456�� )4.208��(0.587) (0.644) (0.549) (0.930) (0.821) (0.806) (0.732) (0.697) (1.124) (1.106)

R2 0.64 0.73 0.78 0.72 0.78 0.49 0.61 0.68 0.61 0.69n 144 144 144 114 114 115 115 115 86 86

a Figures in parentheses are heteroskedastic-consistent (white) standard errors.* p < 0:01:** p < 0:05:*** p < 0:001:


1909

cluded a dummy variable in the equationswhere this ratio took on extreme values (greaterthan two) for any measure. Inclusion of thisdummy variable did not alter the results re-ported below. To be sure, this procedure willnot correct for systematic measurement errorsthat, if present, might bias our parameter esti-mates for the GII coefficients; however, wehave no reason to believe that such systematicbiases characterize the KKZL indices.Table 3 presents results for the full sample of

countries ðn ¼ 144Þ, as well as for a sample thatexcludes OECD members plus Hong Kong andSingapore ðn ¼ 115Þ. The latter sample is re-ferred to as developing and transition econo-mies. Because there were fewer observations forESI, and a more limited sample size, we presenttwo sets of results for each sample: one for thelargest possible sample (n ¼ 144 and 115), butexcluding ESI, and one for a restricted sample(n ¼ 114 and 86) that includes ESI.For the full sample, the first model (1)

presents a simple regression of LnFDI onLnGDP (lagged one period). This model pro-duces a surprisingly high level of explanationfor cross-section estimation ðR2 ¼ 0:64Þ, whichsuggests that GDP acts as a control variable fora variety of economic factors. The coefficienton the GDP term is highly significant andsuggests an elasticity that is not statisticallydifferent from unity. 21

When we add HDI, GII and the interactionof GII and Ln GDP (model 2), or these vari-ables plus ESI (model 4), the explanatorypower of the equations improves. But, the onlyvariable that is statistically significant in bothequations is GII (positive) and its interactionwith Ln GDP (negative, although significantonly at the 10% level in model 4). 22 Thisspecification suggests that while governanceimprovements can attract FDI, they do so at adiminishing rate. That is to say, there are ‘‘di-minishing returns’’ to governance improve-ments, so that the greatest effects will be feltby smaller economies (which are typicallypoorer). Thus, larger and richer countries haveless to gain (at the margin) from governanceimprovements than do smaller and poorereconomies. The smaller impact of governanceimprovements on FDI in larger countries mightbe a statistical artifact of relatively limitedvariation in the governance index across de-veloped countries. To assess this possibility, wecalculated the standard deviations of the GIIvariable across the full sample of countries, aswell as the separate sample of developing andtransition economies. In fact we found that thestandard deviations of GII for the latter sampleis lower than that for the whole sample. In thisregard, the smaller impact of governance onFDI as GDP increases seems a worthy topic forfurther research.

Table 4. Regression coefficients, GII and HDI subindices dependent variable is Ln FDIa

(1) All countries (2) Developing and transition economies

Coefficient(standard error)

R2 Coefficient(standard error)

R2

HDI 2.890� (0.943) 0.67 2.095� (0.740) 0.52Gdp per capitaa 1.674� (0.569) 0.66 1.194� (0.606) 0.50Educationa 2.365� (0.604) 0.67 1.960� (0.616) 0.52Life expectancya 2.142� (0.747) 0.66 1.642� (0.768) 0.50ESIb 0.037� (0.011) 0.69 0.053� (0.016) 0.58GII 0.744� (0.136) 0.73 0.930� (0.182) 0.61Rule of lawc 0.609� (0.161) 0.69 0.640� (0.211) 0.54Voice and accountabilityc 0.669� (0.144) 0.71 0.795� (0.164) 0.58Political instability/violencec 0.649� (0.157) 0.70 0.638� (0.197) 0.56Government effectivenessc 0.748� (0.146) 0.71 0.909� (0.225) 0.57Regulatory burdenc 1.119� (0.154) 0.77 1.105� (0.198) 0.69Graftc 0.595� (0.136) 0.68 0.721� (0.218) 0.57n 144 114

a Figures in parentheses are heteroskedastic-consistent (White) standard errors. Each equation contains an unre-ported constant term and Ln GDP.b Indicates variables that are components of the GII index.c Indicates variables that are components of the HDI index. For this variable n ¼ 114 for the all country sample andn ¼ 86 for the developing and transition economy sample.* p < 0:001:


The HDI and ESI coefficients are not statis-tically significant in these specifications, al-though they are each significant, and positive,when entered individually (with Ln GDP).There are grounds for concern that collinearityamong the independent variables is hamperingreliable estimation of the individual coefficients.Some indirect evidence that the HDI index is astatistically relevant influence on FDI is pro-vided by substituting the educational compo-nent of the HDI index for the HDI variable andthe regulation component of GII for the GIIvariable (Models (3) and (5)). The simple cor-relation between REG and EDUC (0.37) islower than the simple correlation between HDIand GII (0.69), and lower than for any othercombination of the subindices. Both REG andEDUC are less correlated with GII and HDIthan are other subindices. In effect, EDUC andREG can be seen as instrumental variables forthe broader HDI and GII measures. 23

Models 3 and 5 provide improved explana-tory power, with both the EDUC and REGcoefficients positive and statistically significant.This result suggests that the HDI index suffersmore from a collinearity problem with the GIIindex than do the individual components of theHDI index. In any case, the result with respectto EDUC certainly suggests that educationalinfrastructure encourages FDI in the expectedway. As before, the ESI index remains statisti-cally insignificant.To assess the possibility that the stage of a

country�s development, as distinct from acountry�s size, conditions the relationshipbetween FDI and infrastructure measures,we examined a subsample of developing andtransition economies (the full sample lessOECD members as at 1996 plus Hong Kongand Singapore). 24 The results are reported inTable 3, columns 6–10. One difference betweenthese results and those for the full sample isthat interaction term between GII and Ln GDPis never statistically significant in the estimatingequations for the smaller sample of countries,and the non-interactive version passed theRESET test. As a consequence, no interactivespecifications are reported.Although the explanatory power of the

equations estimated for the developing econo-mies is lower than those for the full sample, thecoefficient estimates and levels of significanceare quite similar. The finding that the impact ofpolitical governance does not diminish with sizeamong developing countries suggests that theystand to benefit more at the margin from such

improvements than do richer countries. It isalso noteworthy that we find no evidencethat FDI flows are attracted to developingeconomies where environmental conditions arepoor. 25

We also note that for both samples, a varietyof alternative specifications were estimated withvarious combinations of the main indices (GII,HDI and ESI), and the subindices. We havealready determined that individual componentsof the broader indices such as EDUC and REGare positively related to FDI flows. In order toexamine the potential impact of the othercomponents of the HDI and GII indices, weestimated equations that contain various com-binations of Ln GDP, subindices of HDI andGII, and the interaction of Ln GDP and aninfrastructure subindex. The results indicatethat no interactive terms are statistically signif-icant for any of the subindices, and that when allthe subindices are entered into an equation atonce, only two variables are statistically signif-icant (and positive): education and regulatoryburden. In addition, government effectiveness(GOV) was often nearly statistically significantand positive. In order to give some sense of therelative contributions of these variables, Table 4presents results where variables are individuallyincluded in an equation with Ln GDP. Forcomparative purposes, this table also includesthe GII, HDI and ESI indices.The results confirm the importance of gov-

ernance infrastructure, as measured by GII andthe KKZL indices. The GII index providesmore explanatory power than ESI and HDI asmeasured by the coefficients of determination,and this is true of both samples. In general, theGII index provides more explanatory powerthan do any of its components (except regula-tory burden), while the HDI index does notprovide much advantage over any of its com-ponents. The results also suggest that educationis the most important of the HDI variables(education, per capita GDP and life expec-tancy) as judged by the size of its coefficient,and this is true for both samples. When con-sidering the variables that comprise the gover-nance (GII) index, the regulation coefficient islarger than any of the other coefficients, sug-gesting that open economies with free marketswill attract more FDI than will economies inwhich external and internal competition arediscouraged. This is true regardless of a coun-try�s stage of development, although the effectis marginally weaker for the developing coun-try sample. In both samples, the second most


important of the governance variables is effec-tive government, and this variable is strongerfor developing countries than for developedcountries.Perhaps the most important result to emerge

from Table 4 is that the HDI coefficient and allof its component coefficients are lower in thesample of emerging and transition economies,while the GII coefficient and most of its com-ponent coefficients are higher. This suggeststhat governance is relatively more important todeveloping and transition economies, whilewealth and human capital are relatively moreimportant to developed countries.In summary, our results point in a consistent

direction. Specifically, they confirm the well-established fact that the size of a nationaleconomy strongly conditions how attractivethat location is to foreign investors. They alsostrongly support the notion that governanceinfrastructure is an important direct influenceon FDI, although the influence diminishesas countries become larger. An additional in-ference suggested is that FDI will be morestrongly affected by improvements in politicalgovernance in developing countries than indeveloped countries. Of the governance indi-cators considered, the evidence suggests thatregulatory burden and government effective-ness are the most important determinants ofFDI flows for all countries.There is less reliable evidence regarding other

variables, where issues of collinearity and cau-sality arise. In particular, the HDI is not sig-nificant in the presence of GII. One possiblereason is that the HDI is in fact an outputmeasure which is determined by GII, as sug-gested by Kaufmann et al. (1999b). When weestimate a model with independent variablesthat relatively uncorrelated with each other,however, there is evidence that education levelsare important independent determinants ofFDI flows. Furthermore, there is no evidencethat the impact of this variable decreases withthe size of the country. Finally, we find no evi-dence to suggest that FDI is in any way at-tracted to locations with weaker environmentalregulations or with inferior environmentalquality.

(b) The FDO model

Globerman and Shapiro (1999) argue thatthe same factors encouraging inward FDI in-fluence outward FDI (henceforth FDO), al-though the precise nature of the relationship

between FDO and some of the independentvariables is not clear a priori. On the one hand,factors creating a favorable domestic businessenvironment may both attract foreign capitaland limit capital outflows. In this case, the in-frastructure variables that encourage FDI willdiscourage FDO, and will carry opposite signsin the relevant equations. On the other hand, asdiscussed in an earlier section, the same factorsencouraging foreign-owned MNCs to establishaffiliates in a country may also encouragethe growth of domestically owned MNCs thatthen establish their own affiliates abroad. Inthis context, the infrastructure variables thatdirectly encourage FDI may indirectly encour-age FDO. 26 Thus, an effective domestic gov-ernance infrastructure could well encouragecapital outflows by successful domestic firms.Moreover, the relationships between infrastruc-ture variables and FDO may be more com-plex than relationships between infrastructuremeasures and FDI. For example, alreadyestablished investments by MNCs might becharacterized by relatively large sunk costs. Asa result, factors that have relatively large neg-ative impacts on inward FDI may have muchsmaller impacts on FDO. 27

Clearly, if one is interested in the overallimpact of infrastructure variables on MNCinvestment in a country, one should considerboth FDI and FDO. Hence, we estimated themodel summarized in equation (1) using boththe logarithm of FDO and the logarithm ofFDI minus FDO as dependent variables. 28

Analogous estimates to those reported in Table3 are reported in Table 5. As before, eachequation was subjected to RESET specificationtests resulting in the reported specifications.Each reported equation therefore passed theRESET test.The FDO results are both similar to, and

different from, the FDI results. In both cases, itis clear market size is a crucial determinant ofFDI flows. As economies grow larger, there is atendency for both capital inflows and outflowsto increase. This is true for both samples ofcountries. But, as the net FDI equations for thetotal sample indicate (columns 4 and 5), there isno relationship between Ln GDP and Ln (FDI-FDO), so that the two effects cancel out, onaverage. The same is not true of developing andtransition economies, where the Ln GDP co-efficient is positive and statistically significant(columns 9 and 10), suggesting that for theseeconomies increases in market size result inpositive net capital flows.


Table 5. Regression results, FDI outflows and net FDI flows

All countries Developing and transition economies

Dependent variable is Dependent variable is

Ln FDO Ln (FDI FDO) Ln FDO Ln (FDI FDO)

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

Ln GDP 1.355*** 1.012*** 1.052*** 0.002 )0.309 0.948*** 1.176*** 1.189*** 0.924*** 0.901***(0.092) (0.131) (0.098) (0.204) (0.325) (0.142) (0.147) (0.129) (0.126) (0.094)

Human DevelopmentIndex (HDI)

1.120 )0.141 1.856 )0.862(0.131) (2.734) (1.976) (2.316)

Education Index(EDUC)

1.086 3.713� 0.830 1.163(1.257) (2.243) (1.316) (1.501)

Governance Infrastruc-ture Index (GII)

)1.434 18.489*** )5.433** 0.021(1.150) (2.263) (2.438) (.530)

Regulation Index(REG)

)3.724** 20.859*** )6.446** )2.253(1.582) (3.794) (2.540) (1.391)

Environment Sustain-ability Index (ESI)

)0.010** )0.011*** 0.007 )0.123** )0.009** )0.012** 0.083 0.034(0.004) (0.004) (0.076) (0.073) (0.005) (0.005) (0.069) (0.055)

Ln�GDP� �GII 0.252*** )1.872*** 0.660**(0.093) (0.194) (0.264)

Ln�GDP� �REG 0.538*** )1.992*** 0.818*** 0.331**(0.140) (0.403) (0.261) (0.133)

Constant )9.843*** )7.268*** )7.875*** 5.065 10.704*** )6.348*** )9.378*** )8.976*** )6.743** )6.037***(0.976) (1.205) (1.001) (3.190) (3.766) (1.355) (1.400) (1.234) (2.703) (1.970)

R2 0.60 0.74 0.76 0.56 0.38 0.33 0.44 0.50 0.38 0.43n 98 98 98 98 98 70 70 70 70 70

Figures in parentheses are heteroskedastic-consistent (White) standard errors.

* p < 0:01.** p < 0:05.

*** p < 0:001.


1913

As was the case for FDI, governance infra-structure is an important determinant of FDOfor both samples (columns 2 and 7). In the caseof FDO, however the results suggest that im-provements in the governance infrastructurewill restrict capital outflows for small econo-mies and encourage capital outflows for largereconomies. 29 But, the former effect only holdsfor very small economies, so that for mostcountries the effect is positive. As governanceimproves, most countries should expect to seeincreased capital outflows. For the total sam-ple, the evidence (column 4) suggests that thenet effect is positive for smaller economies, butnegative for larger ones with the latter effectdominating for economies with Ln GDPgreater than 9.7 (compared to a mean of 10.6).When the sample is restricted to developing andtransition economies, the linear specificationwas found to be appropriate, and the GII co-efficient is not statistically significant. Thus, theevidence suggests that for smaller economies,improvements in governance will either resultin positive net capital flows, or will have noeffect on the balance, while for larger econo-mies improvements in governance will likelycreate the conditions for successful domesticfirms to expand abroad.As was the case for FDI, there is no evi-

dence that HDI affects capital outflows. More-over, when we substitute the education index(EDUC) for HDI, and the regulation index(REG) for GII in order to minimize the effectsof multicollinearity (columns 3 and 8), we findthat education is not a significant determinantof FDO in either sample. Investments in edu-cation are therefore not found to result incapital outflows. For the total sample, the neteffects of education on capital flows are positive(column 5), while for developing and transitioneconomies the net effect is zero (column 10).The regulatory burden term performs in

much the same way as the over-all GII index.Movements toward the creation of more openand free markets will limit capital outflows forsmall economies, but not for larger ones, andthe turning point occurs before the mean levelof GDP is reached. The net effect on the totalsample is such that reducing the regulatoryburden will have a net positive effect on FDIflows for small countries, but the overall ef-fect will be negative beyond a critical size thatis reached before the sample mean GDP levelis reached. Thus, relatively small and poorcountries will experience net inflows as a con-sequence of creating more open markets.

Finally, the results in Table 5 provide noevidence to suggest that there is capital flightfrom areas where environmental conditions,including the regulatory environment, are sup-portive of a sustainable environment. Indeed,the relevant ESI coefficient is negative andstatistically significant in the FDO equationsfor both samples. The higher is the index ofenvironmental sustainability, the lower are thecapital outflows. For the most part, the ESIindex is not related to net flows.As was the case for FDI, we examined the

impact of the various indices and subindiceswhen they were entered individually (or withinteractive terms) with Ln GDP. In order toconserve space, we do not present these re-sults. 30 In general, we found that the indi-vidual indices (GII, HDI) provide strongerexplanatory power than any individual subin-dex. We confirmed that education and healthare not significant determinants of FDO flowsfor developing and transition economies, andthat the regulatory burden term is the mostimportant of the GII terms (as judged by R2 forboth samples). For the total sample, it is gen-erally true that raising any of the subindices willcause capital outflows to increase for largeeconomies, and these will also result in lowernet FDI flows for larger countries.In summary, we find fairly strong evidence

suggesting that improvements in governanceinfrastructure will also affect capital outflows,and thus net FDI flows. In general, the ef-fects are strongest for large economies, andare likely most important when liberalization ofthe economic environment is involved. There isone type of liberalization that may not causecapital outflows and that is environmentalregulation. Finally, education infrastructure isnot related to capital outflows and may there-fore have a positive net effect on FDI flows.

6. SUMMARY AND CONCLUSIONS

The purpose of this study is to assess whetherand to what extent the governance infrastruc-ture attributes of national economies influenceFDI flows into and from those economies. Inaddition, we examine the role of other forms ofinfrastructure including human capital and theenvironment. This study therefore focuses on abroad set of indices measured for a relativelylarge sample of countries for the second half ofthe 1990s. Estimation of both FDI inflow and


outflow equations is based upon a symmetricaland parsimonious specification.The results clearly indicate that governance

infrastructure is an important determinantof both FDI inflows and outflows. On balance,for most countries, both inflows and outflowsrespond positively to good governance. Inparticular, good political governance is char-acterized by policies promoting competition onboth a domestic and an international level, aswell as by open and transparent legal and reg-ulatory regimes, and effective delivery of gov-ernment services. These appear to be moreimportant than measures of political voice andpolitical stability, and even the rule of law. Theevidence also suggests that the returns to in-vestments in good governance (in terms of netFDI flows) are greater for developing andtransition economies. 31

The results with respect to human capital(education and health) and environmental sus-tainability are less reliably estimated owing toproblems of multicollinearity and causality. Wedo however find evidence consistent with find-ings of previous studies that investments ineducation are likely to attract FDI. Moreover,there is no indication that the positive impactsof education on FDI are a diminishing functionof real GDP. We also find that such investmentsare not associated with capital outflows, so thatthe net effect of investments in education onFDI flows is likely to be positive. Our findingswith respect to both the human capital andgovernance infrastructure indicators are subjectto the caution that they may reflect the influenceof omitted variables, most notably a measure ofsocial capital. It is also possible that there issome feedback from FDI flows to infrastructurethat would, if significant, contribute to biasedestimates in our regression models.Our results also provide some support for a

claim that initiatives to promote environmentalprotection and remediation encourage, ratherthan discourage, inward FDI. Similarly, wefind no evidence of capital flight from coun-tries pursuing relatively strong environmentalpolicies. While estimation of precise relation-ships between environmental policies and FDIflows is hampered by multicollinearity, statis-tical problems do not obviate rejection of theclaim that governments will engage in an en-vironmental ‘‘race-to-the-bottom’’ in order toattract and retain direct investments. 32 Indeed,our results suggest that weakening environ-mental protection regimes are more likely todiscourage than encourage FDI.

Our findings might lend some support to ar-guments for public investment in infrastructureattributes that have other intrinsic social bene-fits. For example, support for improvements tonational health care systems can be defended ongrounds that improved health care generatespublic goods-type externalities. Increased in-flows of FDI that themselves create spilloverefficiency benefits for host economies offer anadditional source of social benefit to an overallbenefit-cost appraisal of government health careexpenditures. For many developing countries,public goods arguments for improved healthand educational systems, as well as for a cleanerenvironment, may trump considerations abouttheir impacts on FDI. Nevertheless, comple-mentarity between public goods investmentsand inward FDI strengthens the case for suchinvestments in developing countries.Perhaps our most important conclusion

is that political governance matters, and im-proved political governance does not neces-sarily oblige governments to make largeinvestments of taxpayers� money. In this re-gard, our findings reinforce similar conclusionsdrawn in Altomonte (2000) and UNCTAD(1998). Indeed, improved governance might bemore consistent, in many cases, with a smallereconomic and regulatory role for government.As well, any set of policies that broadly pro-motes economic growth will indirectly promoteincreased inward FDI by encouraging a higherlevel of real GDP.Finally, it should be recognized that policies

promoting inward FDI will likely indirectlyencourage increased outward FDI by promot-ing the emergence and growth of successfulhome-country MNCs. But, this developmentshould not necessarily be seen as a ‘‘cost’’ ofinvesting in governance or other types of in-frastructure. Rather, intra- and inter-industryFDI flows among countries facilitate special-ization in production along the same lines asintra- and inter-industry trade. Such special-ization, itself, can be a source of improvedproductivity and real economic growth in hosteconomies.Our results are subject to a number of cave-

ats. In addition to the statistical problemsdiscussed above, it is possible that the parsi-monious specification adopted might result inomitted variable bias. Our specification reliesheavily on GDP as a control variable, andfurther research on the relationship betweenmarket size and other possible determinants ofFDI is warranted.


Another caveat is that the geographical unitof analysis is the nation-state. In fact, theinfluence of infrastructure on the locationdecisions of MNCs might be more geograph-ically circumscribed. For example, MNCs inthe electronics sector will likely have strongpreferences to locate in specific regions of theUnited States regardless of the availability ofinfrastructure in other regions. In this case,

infrastructure comparisons will likely be cir-cumscribed to the few regions that enjoyother necessary pre-conditions. 33 The avail-ability of infrastructure data at the nationallevel dictates the geographical level of ana-lysis. Whether the results would be signi-ficantly different if it were possible to carryout the analysis at subnational levels is un-certain.

NOTES

1. For a broader survey of the literature see Brunetti

(1997).

2. The impact of FDI on host country efficiency arises

from spillover productivity gains. A summary of avail-

able evidence on this phenomenon is provided in

Blomstrom, Kokko, and Globerman (2001).

3. A focus on FDI flows rather than stocks reflects the

fact that FDI data are widely available only for flows.

4. Dunning (1993) offers an extensive review of empir-

ical studies of the determinants of FDI. Globerman and

Shapiro (1999) provide an updated literature review with

particular emphasis on the influence of government

policies.

5. Available studies also identify the specific impact of

government ‘‘corruption’’ on FDI flows. See, for exam-

ple, Wei (2000).

6. See, for example, Nahapiet and Ghoshal (1998),

Pennings, Lee, and van Witteloostuijn (1998); Putnam

(1995) and Sparrowe, Liden, Wayne, and Kraimer

(2001).

7. The data are available at: http://www.worldbank.

org/wbi/governance/datasets.htm#dataset. Further de-

tail is provided in Kaufmann et al. (1999a) (Appendix

A) and Table 1 of this study. The full set of variables

employed in this study and their sources are presented in

Table 2.

8. We did collect data on physical infrastructure, but

these variables were typically highly correlated with

GDP/capita. These measures are discussed below.

9. Kaufmann et al. (1999b) suggest that their gover-

nance measures are important causal determinants of

development outcomes, including GDP/capita and

health status.

10. The indices are all reported in their original scales,

and these scales are also used in the regression analysis.

But, the results are robust to scale transformations,

including logarithmic. The ESI is derived from 22

subindices, a number too large for individual analysis.

11. GII was also measured as the sum of the six

underlying indices. The correlation coefficient between

this measure and the first principal component that we

employ is 0.99.

12. Dunning (1993) identifies an exhaustive list of such

variables and also discusses empirical evidence regarding

their importance. For additional summaries of available

evidence, see Caves (1996) and Globerman and Shapiro

(1999).

13. Some studies identify a near perfect positive

correlation between FDI and GDP across host coun-

tries. See, for example, Morisset�s (2000) study of

African countries.

14. The various sources of agglomeration (or external)

economies are discussed in Krugman (1991).

15. Altomonte (2000) references several studies that

provide ambivalent findings on the relationship between

labor costs and the geographical distribution of FDI.

16. The presumption is that a country will fix its

currency against the major currency (or currencies) in

which its trade and investment flows are most heavily

concentrated. As such, if currency volatility discourages

FDI flows, countries operating fixed exchange rate

regimes should be characterized by more FDI than

those operating floating rate regimes, all other things

constant.

17. The data sources for unreported variables and the

unreported results are available from the authors.


http://www.worldbank.org/wbi/governance/datasets.htm#dataset

http://www.worldbank.org/wbi/governance/datasets.htm#dataset

18. See, for example, Cheng and Kwan (2000); Kumar

(1996); Loree and Guisinger (1995); Mody and Sriniva-

san (1998); Wheeler and Mody (1992); Zhao and Zhu

(2000).

19. These results are available from the authors on

request.

20. This is more likely to be true for developed than for

developing countries. In particular, marked short-run

changes in the ‘‘political’’ environment surrounding FDI

have been identified for certain African countries (Mor-

isset, 2000).

21. This is not however always the case. In specifica-

tions that include other explanatory variables, the

elasticity is at times less than unity. The latter findings

further support the use of FDI, rather than FDI/GDP,

as the preferred specification of the dependent variable.

22. Similar interactive terms were estimated for the

other variables (ESI and HDI), but none were statisti-

cally significant.

23. To be sure, the EDUC variable can also be

interpreted as a conventional human capital variable.

24. For a discussion of this concern, as well as

empirical evidence for Asian MNCs, see Erramilli,

Agarwal, and Kim (1997) and Zhao and Zhu (2000).

25. There is a potential variable measurement problem,

since the ESI index is measured for the year 2000. It

seems unlikely to us however that the values of this

variable changed substantially between, say, the mid-

1990s and 2000. Moreover, this finding is broadly

consistent with those of Smarzynska and Wei (2001)

and Wheeler (2001).

26. There is no notion implied here that FDI is

necessarily good while FDO is bad for a country. Both

flows contribute to an increased specialization of inter-

national production that should improve real incomes

internationally.

27. This argument is developed in Rugman (1990).

28. The sample size is reduced, since a number of

countries recorded FDI but not FDO. This reduction

allowed us to use a single sample with ESI in all

estimates. Ln (FDI-FDO) was calculated by separating

countries with net inflows and outflows, and assigning a

negative value to the latter.

29. The point at which Ln GDP encourages capital

outflows occurs where Ln GDP ¼ 5:6 for the total

sample, and Ln GDP ¼ 8:2 for the sample of developing

and transition economies. The respective means (stan-

dard deviations) are 10.4 (2.7) and 9.6 (1.6). In both

cases it is only the very smallest economies for which

improvements in governance do not result in capital

outflows.

30. The results are complicated by the fact that some of

the preferred specifications include an interactive term

and some do not. In general, specifications for the total

sample include interactive terms, while those for the

sample of developing and transition economies do not.

31. This result is consistent with an interpretation

suggested by Altomonte (2000). Namely, that the

‘‘quality’’ of a host country�s legal framework is posi-tively related to expected future rates of economic

growth. Hence, FDI will be attracted to countries with

higher quality legal frameworks because of the superior

economic performance that those frameworks encour-

age. If the economic benefits of higher quality legal

frameworks are subject to diminishing returns, the

linkage between FDI and political governance might

be nonlinear.

32. Thomsen (2000) discusses these and related objec-

tions to international investment.

33. The relevant notion here is that firms are attracted

to regions that are home to ‘‘clusters’’ of firms engaged

in the same or closely related activities. In the United

States, firms engaged in micro-electronics-related activ-

ities are disproportionately located on the East and West

Coasts.

REFERENCES

Altomonte, C. (2000). Economic determinants andinstitutional frameworks: FDI in economies in tran-sition. Transnational Corporations, 9(2), 75–106.

Bevan, A., & Estrin, S., (2000). The determinants offoreign direct investment in transition economies.

Centre for New and Emerging Markets, DiscussionPaper Number 9, London Business School, London.

Blomstrom, M., Kokko, A., & Globerman, S. (2001).The determinants of host country spillovers fromforeign direct investment: a review and synthesis of


the literature. In N. Pain (Ed.), Inward investment,technological change and growth (pp. 34–65). Basing-stoke: Palgrave.

Brunetti, A. (1997). Political variables in cross-countrygrowth analysis. Journal of Economic Surveys, 11(2),163–190.

Caves, R. E. (1996). Multinational enterprise and eco-nomic analysis. Cambridge: Cambridge UniversityPress.

Chen, D. (2000). The marginal effective tax rate: the onlytax rate that matters in capital allocation. C.D. HoweInstitute Backgrounder, 1–11.

Cheng, L. K., & Kwan, Y. K. (2000). What are thedeterminants of the location of foreign direct invest-ment. the Chinese experience. Journal of Interna-tional Economics, 379–400.

Dawson, J. W. (1998). Institutions, investment andgrowth: new cross-country and panel data evidence.Economic Inquiry, 36, 603–619.

Dunning, J. H. (1993). Multinational enterprises and theglobal economy. Wokingham, England: Addison-Wesley Publishing Company.

Erramilli, M. K., Agarwal, S., & Kim, S. S. (1997). Arefirm-specific advantages location-specific too? Jour-nal of International Business Studies, Fourth Quarter,735–757.

Globerman, S., & Shapiro, D. (1999). The impact ofgovernment policies on foreign direct investment: theCanadian experience. Journal of International Busi-ness Studies, 30(3), 513–532.

Hall, R., & Jones, C. I. (1999). Why do some countriesproduce so much more output per worker thanothers. Quarterly Journal of Economics, 114(1), 83–86.

Humphrey, J., & Schmitz, H. (1998). Trust and inter-firm relations in developing and transition econo-mies. Journal of Developing Studies, 34(4), 32–61.

Kaufmann, D., Kraay, A., & Zoido-Lobaton, P.(1999a). Aggregating governance indicators. WorldBank Working Paper #2195, available at: http://www.worldbank.org/wbi/governance.

Kaufmann, D., Kraay, A., & Zoido-Lobaton, P.(1999b). Governance matters. World Bank WorkingPaper #2196, available at: http://www.worldbank.org/wbi/governance.

Keefer, P., & Knack, S. (1997). Why don�t poor countriescatch up. a cross-national test of an institutionalexplanation. Economic Inquiry, 35, 590–602.

Knack, S., & Keefer, P. (1995). Institutions andeconomic performance: cross-country tests usingalternative institutional measures. Economics andPolitics, 7, 207–227.

Knack, S., & Keefer, P. (1997). Does social capital havean economic payoff ? Quarterly Journal of Economics,112(4), 1251–1288.

Kobrin, S. J. (1976). The environmental determinants offoreign direct investment: an ex-post empirical ana-lysis. The Journal of International Business Studies,7(2), 29–42.

Krugman, P. (1991). Geography and trade. Leuven,Belgium: Leuven University Press.

Kumar, N. (1996). Intellectual property protection,market orientation and the location of overseas

R&D activities by multinational enterprises. WorldDevelopment, 24(4), 673–684.

List, J. A. (2001). US county-level determinants ofinbound FDI: evidence from a two-step modifiedcount data model. International Journal of IndustrialOrganization, 19, 953–973.

Loree, D. W., & Guisinger, S. (1995). Policy and non-policy determinants of US equity foreign directinvestment. Journal of International Business StudiesSecond Quarter, 281–299.

Lucas, R. E. (1990). Why doesn�t capital flow from richto poor countries? The American Economic Review,80(2), 92–96.

Mody, A., & Srinivasan, K. (1998). Japanese and USfirms as foreign investors: do they march to the sametune? Canadian Journal of Economics, 31(4), 778–800.

Morisset, J. (2000). FDI in Africa: policies also matter.Transnational Corporations, 9(2), 107–126.

Nahapiet, J., & Ghoshal, S. (1998). Social capital,intellectual capital and the organizational advantage.The Academy of Management Review, 23(2), 242–266.

OECD (2001). The well-being of nations: the role ofhuman and social capital. OECD, Paris.

Pennings, J. M., Lee, K., & van Witteloostuijn, A.(1998). Human capital, social capital and firmdissolution. Academy of Management Journal,41(4), 425–440.

Peterson, M., Malhota, N. K., & Wagner, J. (1999).Country quality of life and foreign direct investmentdecisions. Global Outlook, 11(1), 51–62.

Putnam, R. D. (1995). Bowling alone: America�s declin-ing social capital. Journal of Democracy, 6, 65–78.

Roll, R., & Talbott, J. (2001). Why many developingcountries just aren�t. San Diego: Mimeo: WorldDevelopment Inc.

Rugman, A. M. (1990). Multinationals and Canada-United States free trade. Columbia: University ofSouth Carolina Press.

Smarzynska, B. K., & Wei, S. J. (2001). Pollution havensand foreign direct investment: dirty secret of popularmyth.National Bureau of Economic Research Work-ing Paper No. W8465, NBER, Cambridge, MA.

Sparrowe, R. T., Liden, R. C., Wayne, S. J., & Kraimer,M. L. (2001). Social networks and the performanceof individuals and groups. Academy of ManagementJournal, 44(2), 316–325.

Stevens, G. V. G. (2000). Politics, economics andinvestment: explaining plant and equipment spend-ing by US direct investors in Argentina, Brazil andMexico. Journal of International Money and Finance,19(2), 115–135.

Thomsen, S. (2000). Investment patterns in a longer-termperspective. Mimeo, Organization for Economic Co-operation and Development, Working Paper onInternational Investment Number 2000/2, Paris.

Tuman, J., & Emmert, C. (1999). Explaining Japaneseforeign direct investment in Latin America, 1979–1992. Social Science Quarterly, 80(3), 539–555.

United Nations Conference on Trade and Development(UNCTAD) (1998). World investment report 1998:trends and determinants. New York: United Nations.

United Nations (various years). Statistical Yearbook.New York: UN.


http://www.worldbank.org/wbi/governance



UNDP (various years). Human Development Index. NewYork: UNDP.

Vining, A. R., & Weimer, D. L. (2001). Criteria forinfrastructure investment: normative, positive andprudential perspectives. In A. R. Vining & J.Richards (Eds.), Building the future: issues in publicinfrastructure in Canada (pp. 131–165). Toronto:C.D. Howe Institute.

Wei, S. J. (2000). How taxing Is corruption on interna-tional investors? Review of Economics and Statistics,82(1), 1–11.

Wheeler, D., & Mody, A. (1992). International invest-ment locational decisions––the case of US firms.Journal of International Economics, 33, 57–76.

Wheeler, D. (2001). Racing to the bottom? Foreigninvestment and air pollution in developing countries.World Bank Working Paper no. 2524, World Bank,Washington DC.

Zhao, H., & Zhu, G. (2000). Location factors andcountry-of-origin differences: an empirical analysisof FDI in China. Multinational Business Review(Spring), 60–73.


1-s2.0-s0305750x02001109-main

Documents

forms of infrastructure

capital returns

deneits investment environment

capital ows fromrich

human capital

foreign investors

knack keefer

economic growth